INDEX

Explanations

depolarization, cstring, allclose, redraw

The neuron fires on occurrences of the substring “depolar” (as in “depolarization”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

-2.16

-2.14

 THESE

-2.11

 bénévol

-2.11

☝

-1.98

-1.88

 énergie

-1.88

these

-1.87

-1.84

 иметь

-1.83

POSITIVE LOGITS

囪

2.08

ation

2.06

傒

2.03

鲧

2.02

 übrigen

1.98

 οποία

1.98

 favored

1.97

鹮

1.97

艄

1.93

꯳

1.92

Activations Density 0.001%